to quantify the appropriateness of pain management in nursing homes, Hutt and colleagues (2006) calculated a mean score of 66 percent of optimal pain management in residents of 12 nursing homes in Colorado (Hutt et al., 2006). Fewer than half of the residents with predictably recurrent pain had prescriptions for scheduled pain medication, and only 40 percent with neuropathic pain were on an appropriate analgesic adjuvant.

Overutilization of medication, another indicator of inappropriate prescribing, was demonstrated in a retrospective chart review of the use of histamine-2 (H₂) receptor blocker therapy among 711 residents in one academic nursing home (Gurwitz et al., 1992). H₂ blocker therapy was used for unsubstantiated indications in 41 percent of the 110 residents receiving this category of drugs.

For the purposes of this study, the committee examined medication error rates in six different settings within the ambulatory care domain: (1) the interface between care settings, for example, from hospital care to outpatient clinic; (2) the ambulatory clinic; (3) the community or mail order pharmacy; (4) the home care setting; (5) the self-care setting; and (6) the school setting.

It is believed that medication errors and ADEs occur frequently in the interfaces between care settings, particularly after hospital discharge, yet the committee could find only two studies estimating error rates for such transitions (see Table C-10). In one study, a total of 42 (49 percent) patients who were discharged from the hospital and received continuing care from their primary care physicians experienced at least one medication error within 2 months of hospital discharge (Moore et al., 2003). In the other study, 45 (11 percent) of the 400 patients discharged from a general medicine service